Search

Medical researchers are legally required to protect patients' privacy by removing personally identifiable information from medical records before sharing the data with other researchers. We present an evaluation of methods for computer-assisted removal and replacement of protected health information (PHI) from free-text nursing notes collected in the intensive care unit as part of the MIMIC II project. A semiautomated method was developed to allow clinicians to highlight PHI on the screen of a tablet PC and to compare and combine the selections of different experts reading the same notes. An analysis of the performance of three human expert de-identifiers and of an automated system is presented. Expert adjudication demonstrated that inter-human variability was high, with few false positives and many false negatives. The sensitivity of human experts working alone ranged from 0.63 to 0.93, with an average of 0.81, and the average positive predictive value was 0.98. An algorithm generated few false negatives but many false positives. Its sensitivity was 0.85, but its positive predictive value was only 0.37. The de-identified database of nursing notes was re-identified with realistic surrogate (but unprotected) dates, serial numbers, names, and phrases to provide a gold standard database of over 2600 notes (approximately 340,000 words) with over 1700 instances of PHI. This reference gold standard database of nursing notes and the Java source code used to evaluate algorithm performance will be made freely available on Physionet in order to facilitate the development and validation of future de-identification algorithms.

We have developed a new technique, which may enable a more accurate, complete characterization of the circulatory system, including local or global hydrodynamic phenomena, using multiple measurements from several anatomic locations and/or multiple modalities. This technique, Laguerre model blind system identification (LMBSI), uses a Laguerre function series expansion to provide a compact but complete quantitative description of the distinct behavior of two or more circulatory waveforms. LMBSI identifies a set of five parameters per channel plus one common parameter that can be treated as a feature vector and used to predict cardiovascular parameters of interest. Standard statistical techniques can be used to extract information from that compact feature vector. In this paper, multiparameter regression is used to predict cardiac output, using two separate arterial pressure waveforms and the LMBSI algorithm. This serves as a proof-of-principle that two distinct circulatory waveforms, with LMBSI, can be used to characterize the circulatory system. In the future, this technique might be applied to noninvasive circulatory measurements.

An adaptive approach is proposed for the problem of left ventricular ejection time (LVET) estimation using peripheral pressure waveform signals. The proposed algorithm, which makes use of 2 peripheral pressure measurements, makes it possible to adaptively estimate the LVET in response to different cardiovascular physiologic states. The algorithm builds on features obtained from global and branch-specific characterization of the cardiovascular circulation as well as waveform features to dramatically improve the accuracy of LVET estimation. The performance of the proposed approach is evaluated with respect to its heart-rate-based conventional counterpart, which shows approximately 40% improvement of estimation accuracy in terms of Rsup2/supvalues#9633; from 0.6655 for the conventional waveform-based approach to 0.9222 for the proposed approach.

All personally identifiable information must be removed from patient medical records before the data can be shared with other researchers. We present an automated method of removing protected health information (PHI) from free-text nursing notes taken from a U.S. hospital. We have previously shown that one clinician can locate PHI in nursing notes with an average sensitivity of 0.81, and for teams of two clinicians the sensitivity is 0.94. Our method uses lexical look-up tables, regular expressions, and simple heuristics to locate PHI with an overall sensitivity of 0.92 (0.98 for names, 0.96 for dates), which is significantly better than the average sensitivity of a single human. The algorithm has a positive predictive value of only 0.44, so additional software was developed to allow the user to review the terms identified as PHI and manually eliminate false positives. The algorithm is open-source and will be made freely available on PhysioNet together with a re-identified corpus of nursing notes

Cardiac output (CO) estimation using arterial blood pressure (ABP) waveforms has been an active area of physiology research over the past century. However, the effectiveness of the estimators has not been extensively studied in a clinical setting. In this paper, we evaluate 11 well-known CO estimators using clinical radial ABP waveforms from the multi-parameter intelligent monitoring for intensive care II (MIMIC II) database, using thermodilution CO (TCO) as reference for comparison. We compare estimations to 988 TCO measurements in 84 patients, totaling 165 hours of ABP waveforms sampled at 125 Hz. As a necessary step for producing absolute CO estimates, we also present 3 methods of calibrating the estimators, each tailored towards a different use model. The results show that the standard deviation of error between TCO and the best CO estimators is approximately 1 L/min for absolute CO estimates. For relative estimates without calibration, the best CO estimator has 18% error at 1 standard deviation

A new tool has been developed for characterizing both the global and local dynamic behavior exhibited by peripheral circulatory waveforms. The signals representing these waveforms can be derived from a variety of different cardiovascular sensor modalities measuring changes in pressure, flow, or volume. The correlation that exists between two peripheral circulatory waveform measurements taken at different locations allows this new signal-processing algorithm to identify two compact, low order dynamic models in real-time. These models identify the distinct dynamic behavior of the circulatory paths traveled by the pulsatile waveforms using output measurements alone. This new algorithm is based on a multi-channel blind system identification technique that has been reformulated to use a Laguerre basis function series expansion. Additionally, a new deconvolution algorithm has been developed to allow estimation of the unknown common input using the identified Laguerre models, which are generally inversely unstable. This new approach has been shown to provide accurate identification of vascular hemodynamics when applied to experimental data derived from a swine.Copyright © 2004 by ASME

We address both technical and clinical issues of wearable biosensors (WBS). First, design concepts of a WBS are presented, with emphasis on the ring sensor developed by the author's group at MIT. The ring sensor is an ambulatory, telemetric, continuous health-monitoring device. This WBS combines miniaturized data acquisition features with advanced photoplethysmographic (PPG) techniques to acquire data related to the patient's cardiovascular state using a method that is far superior to existing fingertip PPG sensors. In particular, the ring sensor is capable of reliably monitoring a patient's heart rate, oxygen saturation, and heart rate variability. Technical issues, including motion artifact, interference with blood circulation, and battery power issues, are addressed, and effective engineering solutions to alleviate these problems are presented. Second, based on the ring sensor technology the clinical potentials of WBS monitoring are addressed.

A clinical method for monitoring cardiac output (CO) should be continuous, minimally invasive, and accurate. However, none of the conventional CO measurement methods possess all of these characteristics. On the other hand, peripheral arterial blood pressure (ABP) may be measured reliably and continuously with little or no invasiveness. We have developed a novel technique for continuously monitoring changes in CO by mathematical analysis of a peripheral ABP waveform. In contrast to the previous techniques, our technique analyzes the ABP waveform over time scales greater than a cardiac cycle in which the confounding effects of complex wave reflections are attenuated. The technique specifically analyzes 6-min intervals of ABP to estimate the pure exponential pressure decay that would eventually result if pulsatile activity abruptly ceased (i.e., after the high frequency wave reflections vanish). The technique then determines the time constant of this exponential decay, which equals the product of the total peripheral resistance and the nearly constant arterial compliance, and computes proportional CO via Ohm's law. To validate the technique, we performed six acute swine experiments in which peripheral ABP waveforms and aortic flow probe CO were simultaneously measured over a wide physiologic range. We report an overall CO error of 14.6%.

The effects of traumatic brain injury on family dynamics and finances have not been extensively studied. We designed a 30-item survey to determine the effects of the injury on the child, parent(s), and siblings, and whether parents were retrospectively content with the decisions related to aggressiveness of care.A questionnaire written on a sixth-grade reading level was mailed to 46 families of children who survived a traumatic brain injury resulting in an initial Glasgow Coma Scale scoreor = 7. This is a descriptive study without statistical analysis. Human Studies Committee review and informed consent were obtained.Thirty-two of 46 (69.6%) surveys were completed by mail or telephone. Mean patient age at time of injury was 8.7 years, mean time since injury was 3 years, and mean Glasgow Coma Scale score was 3.7. Approximately one third of children have disabilities related to education, socialization, and/or self-care skills; require multiple health care visits each month; and require prescription medications. Over one third of families were impacted in a moderate to profoundly negative way; 30% of families reported a deterioration in finances or loss of job; and 16% reported a worsening of adult relationships. In 13 of 32 cases, modification of current housing or new housing was required to facilitate home care. Siblings were adversely affected in approximately 16 of 28 families, exhibiting behavioral problems, increased fear, and withdrawal from the injured child. Only 1 of 32 families stated that they would have considered less aggressive treatment, even if it led to the child's death.Traumatic brain injury of a child had a negative impact on family dynamics and/or sibling behavior in more than one third of families. Despite this, most families would not limit or withdraw care if they could reconsider decisions regarding treatment. This information can be used to counsel families.

This paper presents a modal decomposition approach to continuous cardiac output (CO) monitoring via multi-channel blind system identification (BSI). The presented method estimates CO from multiple noninvasive sensors located at different branches of the systemic circulatory system.Copyright © 2002 by ASME

A novel method for diagnostic monitoring of the cardiovascular system is presented, which combines peripheral measurements obtained continuously and non-invasively with a multi-channel blind system identification technique.Copyright © 2002 by ASME

Trauma is the leading cause of death in children ages 1 to 21 years of age. Traumatic brain injury (TBI) poses a high risk of both morbidity and mortality within the subset of pediatric trauma patients. Numerous adult studies have shown that coagulopathy is commonly observed in patients who have sustained trauma and that the incidence is higher when there is TBI. Previously, it was thought that coagulopathy related to trauma was dilutional (i.e. due to replacement of red cells and platelets without plasma) but more recent studies show that the coagulopathy in trauma is early and likely independent of transfusion therapy. Additionally, abnormal coagulation studies (PT, PTT, INR, platelet count, fibrinogen, and D-dimer) following TBI are associated with increased morbidity and mortality in adults. Although coagulopathy after traumatic brain injury in adults is well documented, the pediatric literature is fairly sparse. A recent study by Hendrickson et al in 2008 demonstrated that coagulopathy is both underestimated and under-treated in pediatric trauma patients who required blood product replacements. Here we present the results of a retrospective pilot study designed to assess coagulopathy in the pediatric TBI population. We analyzed all children admitted to our facility with TBI from January 2012 to December 2013. Patients were excluded if they had underlying diseases of the hemostatic system. All patients had baseline characteristics measured including: age, sex, mechanism of injury, Glasgow Coma Scale (GCS), injury severity score (ISS), initial complete blood count, DIC profile, hematological treatments including transfusions, ICU and hospital length of stay, ventilator days and survival status. Coagulation studies were defined as "abnormal" when they fell outside the accepted reference range of the pediatric hospital laboratory (PT 12.6-15.9, PTT 23.6-42.1 seconds, fibrinogen < 180 mg/dL units, platelets < 185 103/mL and hemoglobin < 11.5 g/dL). Survival was measured as survival at 30 days from admission or last known status at hospital discharge. One hundred and twenty patients met the inclusion criteria of the study and all were included in outcome analysis. Twenty-three of the 120 patients died (19.2%). Logistic regression analysis was used to compare survivors and non-survivors and baseline demographic data showed no difference in age or weight between the two groups with p-values of 0.1635 and 0.1624, respectively. Non-survivors had a higher ISS (30.26 vs 20.92, p-value 0.0004) and lower GCS (3 vs 5.8, p-value 0.0002) compared to survivors. Univariate analysis of coagulation studies to mortality showed statistically significant odds-ratios for ISS (OR 1.09, 95% CI 1.04-1.15), PT (OR 5.91, 95% CI 1.86-18.73), PTT (OR 6.48, 95% CI 2.04-20.52) and platelets (OR 5.63, 95% CI 1.74 – 18.21). Abnormal fibrinogen levels were not predictive of mortality (OR 2.56, 95% CI 0.96-6.79). These results are summarized in Table 1. Our results demonstrate that, consistent with adult studies, abnormal coagulation studies are also associated with increased mortality in pediatric patients. Higher injury severity scores and lower GCS scores are also predictive of mortality. Taken together, these results suggest that possible early correction of coagulopathy in severe pediatric TBI patients could improve outcomes for these patients. Table 1. OR 95% CI p-value ISS 1.09 1.04—1.15 .0009 PT > 15.9 sec 5.91 1.86—18.73 0.0026 PTT > 42.1 sec 6.48 2.04—20.52 0.0015 Fibrinogen < 180 mg/dL 2.56 0.96—6.79 0.0597 Platelets < 185 x 103/mL 5.63 1.74—18.21 0.0040 Disclosures No relevant conflicts of interest to declare.

muchofthedatacontainedinEHRs are stored as free-text documents and are difficult to access and review efficiently by care providers, which may lead to redundant or inappropriate utilization of health care resources.Inresponsetothisdifficulty, a team of clinicians and software developers within our Department of Radiology’s informatics division has leveraged advanced EHR technologytodevelopasearchenginefor theEHRknownasQueriablePatient Inference Dossier (QPID) [8]. QPID

We report a simple explant technique to isolate and propagate endothelial cells from bovine cerebral arteries. The endothelial nature of the cells was confirmed by the presence of Factor VIII/von Willebrand antigen, the ability to phagocytize low-density lipoprotein, and the ability to be induced to express E-selectin. The lack of expression of the CD11c antigen and the absence of smooth muscle alpha-actin immunofluorescence suggested that the cultures were not contaminated with macrophages or smooth muscle cells, respectively. This technique yields pure cerebral arterial endothelial cell cultures, which will be of value for in vitro investigation of cerebrovascular physiology and disease processes.

Purpose: Measurement error and transient variability affect vital signs. These issues are inconsistently considered in published reports and clinical practice. We investigated the association between major hemorrhagic injury and vital signs, successively applying analytic techniques that excluded unreliable measurements, reduced transient variation, and then controlled for ambiguity in individual vital signs through multivariate analysis. Methods: Vital sign data from 671 adult prehospital trauma patients were analyzed retrospectively. Computer algorithms were used to identify and exclude unreliable data and to apply time averaging. An ensemble classifier was developed and tested by cross-validation. Primary outcome was hemorrhagic injury plus red cell transfusion. Areas under receiver operating characteristic curves (ROC AUCs) were compared by the test of DeLong et al. Results: Of initial vital signs, systolic blood pressure (BP) had the highest ROC AUC of 0.71 (95% confidence interval, 0.64-0.78). The ROC AUCs improved after excluding unreliable data, significantly for heart rate and respiratory rate but not significantly for BP. Time averaging to reduce temporal variability further increased AUCs, significantly for BP and not significantly for heart rate and respiratory rate. The ensemble classifier yielded a final ROC AUC of 0.84 (95% confidence interval, 0.80-0.89) in cross-validation. Conclusions: Techniques to reduce variability in vital sign data can lead to significantly improved diagnostic performance. Failure to consider such variability could significantly reduce clinical effectiveness or confound research investigations.

Vascular supply plays a significant role in the management of skull base tumors. The diagnosis is aided by contrast-enhanced imaging and angiographic techniques, and embolization procedures are used to devascularize certain lesions. The degree of surgical technical difficulty is strongly influenced by the degree of tumor vascularity. Although the importance of this blood supply is clearly understood, the mechanism involved in developing a system of tumor-perfusing vessels is yet to be defined. The development of a vascular network, or angiogenesis, is an important event in allowing tumor proliferation to progress beyond small clusters of cells. Basic fibroblastic growth factor (bFGF) is an especially attractive candidate as an angiogenic growth factor because of its ability to stimulate processes that are characteristic of angiogenesis in vitro. Tumors that involve the meninges may have the ability to liberate normally stored bFGF, which may, in turn, induce new vessel formation for continued tumor proliferation. An immunohistochemical analysis of rodent and bovine meninges to study this phenomenon is described. The dura, arachnoid, and their associated vessels are shown clearly to contain this growth factor. Ultimately, an adjuvant therapy based on the inhibition of angiogenesis may provide a reasonable alternative to aggressive surgical approaches in skull base tumors that are incompletely resectable.

ObjectiveDuring initial assessment of trauma patients, metrics of heart rate variability (HRV) have been associated with high-risk clinical conditions. Yet, despite numerous studies, the potential of HRV to improve clinical outcomes remains unclear. Our objective was to evaluate whether HRV metrics provide additional diagnostic information, beyond routine vital signs, for making a specific clinical assessment: identification of hemorrhaging patients who receive packed red blood cell (PRBC) transfusion.MethodsAdult prehospital trauma patients were analyzed retrospectively, excluding those who lacked a complete set of reliable vital signs and a clean electrocardiogram for computation of HRV metrics. We also excluded patients who did not survive to admission. The primary outcome was hemorrhagic injury plus different PRBC transfusion volumes. We performed multivariate regression analysis using HRV metrics and routine vital signs to test the hypothesis that HRV metrics could improve the diagnosis of hemorrhagic injury plus PRBC transfusion vs routine vital signs alone.ResultsAs univariate predictors, HRV metrics in a data set of 402 subjects had comparable areas under receiver operating characteristic curves compared with routine vital signs. In multivariate regression models containing routine vital signs, HRV parameters were significant (P < .05) but yielded areas under receiver operating characteristic curves with minimal, nonsignificant improvements (+0.00 to +0.05).ConclusionsA novel diagnostic test should improve diagnostic thinking and allow for better decision making in a significant fraction of cases. Our findings do not support that HRV metrics add value over routine vital signs in terms of prehospital identification of hemorrhaging patients who receive PRBC transfusion.